Chronic stress, with adverse effects on the immune system and on inflammatory response, may be an important cause of preterm birth and associated racial disparities in birth outcomes. However, an association between maternal stress and preterm birth has been inconsistently reported in the literature. One reason may be the use of multiple and varied surveys, tools, and instruments to measure self-reported stress in pregnancy. Using item response theory (IRT) modeling, we propose to develop a single, comprehensive, validated scale to measure self- reported stress in pregnancy - or an optimized stress in pregnancy scale. The scale will be derived from the responses of 2,006 women to multiple self-reported stress surveys in the Pregnancy, Infection and Nutrition (PIN) Study database. Specifically, IRT characterizes each item according to its location on the concept's continuum (e.g., stress severity) and its ability to discriminate individuals along this continuum. Having this item-level data permits the identification of the best items for an optimized scale. We propose to evaluate the association between self-reported stress, measured by the optimized stress in pregnancy scale and CRH and cortisol levels and between self-reported and preterm birth in women in the PIN Study. An optimized scale of self-reported stress in pregnancy is essential for future studies to be able to examine causal relationships between self-reported stress, biologic markers of stress, and preterm birth. If chronic maternal stress can be shown to be predictive of preterm birth, an optimized stress in pregnancy scale could serve as a screening tool of high-risk pregnant women and as an evaluation tool of treatment strategies to mitigate stress during pregnancy and to reduce preterm delivery and disparities in preterm birth. PUBLIC HEALTH RELEVANCE: The goal of the proposed research is to create a more efficient and effective scale to measure self-reported stress in pregnancy by applying Item Response Theory modeling to the responses to multiple stress surveys already collected from 2,006 pregnant women in the Pregnancy, Infection and Nutrition (PIN) Study and, then, to investigate associations between self-reported stress, measured by the optimized stress scale, and stress biomarkers and preterm birth for women in the PIN Study. If chronic maternal stress can be shown to be predictive of preterm birth, then an optimized stress in pregnancy scale could serve as a screening tool of high-risk pregnant women and as an evaluation tool of treatment strategies to mitigate stress during pregnancy and to reduce preterm delivery and disparities in preterm birth.